Economics revolves around the study of benefits and costs. The goal of any economic action is to maximize net benefits (Total Benefits minus Total Costs). That maximization occurs when marginal benefits (the incremental benefit achieved from one additional unit) is equal to the marginal cost (the incremental cost achieved from one additional unit).
Whenever there is discussion in the economic world on different policies (for example, minimum wage, government stimulus, or protectionist tariffs), we spend much time discussing the costs and the benefits.* And, indeed, there are some studies that find that the estimated benefits for some government intervention will outweigh its estimated costs.
So, why then do I oppose these measures if the research may show a gain in benefits? The reason why is simple: those who benefit are not necessarily those paying the cost (and, as is often the case, those for whom the benefit is intended don’t receive it).
Let’s take, for example, minimum wage. As a casual search of this blog shows, I vehemently oppose minimum wage legislation in all its forms. Minimum wage is designed to help the poorest workers earn a little more. However, the beneficiaries of such legislation tends to be current minimum wage workers who are relatively better off and more secure in their skill set: they tend to be white, middle-class teenagers who work minimum wage to earn a few bucks. Those who tend to pay the costs of minimum wage tends to be current minimum wage workers with fewer skills: they tend to be minorities, immigrants, or less-educated folks. In the discussion of minimum wage, it fails its objective to adequately help the poorest workers. Additionally, those who benefit are not the same as those who pay the costs.
Another example is protectionist tariffs. The beneficiaries are clearly designated: the companies (and their stockholders, profit holders, and employees) protected. The costs are harder to determine, but they are there: the purchasers who must now pay a higher price for the same goods. Again, we see the costs and the benefits do not accrue to the same person.
This is where the discussion of costs and benefits in an aggregate sense runs into issues. The discussion of costs and benefits, MC = MB and all that, that we economists discuss in our textbooks and undergraduate classrooms focuses on the costs and benefits accrued to the single economic actor (the individual, the firm, or the institution). In which case, the economic actor pays both the cost and enjoys the benefit. He/She/It can make a rational decision.** However, when aggregating, the benefits do not necessarily accrue to the same actor as the costs and it can easily lead to poor decisions that make people worse off even if the benefits outweigh the costs!
The TL;DR version of this post is this: At an aggregate level, looking at the total benefits and total costs is meaningless. One needs to look at who benefits and who pays. By focusing only on the aggregate and not on the micro level, one may conceivably design a policy that increases total net benefits but those benefits accrue to someone for whom the policy wasn’t intended.
*I say “economic world” because in the political world, these items are often treated as costless.
**Rational does not mean perfect, by the way. The fact that people may make bad decisions is not a case for government intervention or a breakdown in the market theory or its underlying assumptions.